Patient-Reported Outcomes Beat NYHA Class for Tracking Gains in HFrEF

For patients with heart failure with reduced ejection fraction (HFrEF), patient-reported outcomes are a more-sensitive measure of health changes over time than the old standby, New York Heart Association (NYHA) class, according to an analysis of patients from the CHAMP-HF study.

By clinician-assigned NYHA assessment, 65% of the cohort had no improvement in class over 12 months of follow-up. When the patient-reported Kansas City Cardiomyopathy Questionnaire Overall Summary Score (KCCQ-OS) was used, however, 75% showed a clinically meaningful change. Improvements on the KCCQ-OS, but not those in NYHA class, were linked to a reduction in mortality.

To TCTMD, senior investigator Gregg Fonarow, MD (University of California, Los Angeles), said the discrepancy shows that in terms of aiding physician interpretation of individual health status over time among patients with HFrEF, the KCCQ-OS is a far more sensitive instrument and, ultimately, one that is more clinically actionable.

“The bottom line is that [the KCCQ-OS] provides an additional important piece that is part of routine serial assessment of patients with heart failure,” he said. “We should be looking towards ways of integrating serial collection, interpretation, and acting upon patient-reported outcome measures.”

Writing in an accompanying editorial, Paul A. Heidenreich, MD (VA Palo Alto Health Care System, CA), notes that the KCCQ can be completed electronically by patients in about 7 minutes in the waiting room. It can then go directly to the physician for review before seeing the patient.

“Because patients are now accustomed to previsit administrative questions and postvisit satisfaction  surveys, a previsit electronic survey will not be unusual,” Heidenreich writes. “Such surveys can be incorporated into contactless electronic check-in now common during the coronavirus disease 2019 pandemic.”

KCCQ-OS Change Correlates With Less Death, Hospitalization

For the analysis, published online March 24, 2021, ahead of print in JAMA Cardiology, Fonarow and colleagues led by Stephen J. Greene, MD (Duke Clinical Research Institute, Durham, NC), examined data on 2,872 patients from 145 cardiology practices who were enrolled in the CHAMP-HF registry between December 2015 and October 2017. At baseline, the majority of patients (59.5%) were categorized as NYHA class II. Nearly 40% of patients had baseline KCCQ-OS scores in the best health -status category (75-100) and 5.6% were in the worst health category (0-24).

When baseline concordance/discordance between NYHA class and KCCQ-OS was measured, 37.8% had concordance, 52% had mild discordance by 1 level, and 10.2% had moderate to severe discordance by 2 to 3 levels. Most of the cases of discordance involved worse NYHA class (68.4%) rather than worse KCCQ-OS score (31.6%). Discordance with the EuroQoL 5-dimensions (EQ-5D) utility index and the EQ-5D visual analog scale was more common for NYHA class than KCCQ-OS (75.2% vs 64.6%). For both NYHA class and KCCQ-OS, discordance with EQ-5D scales was primarily due to better EQ-5D status (ie, mobility, self-care, usual activity, pain or discomfort, and anxiety or depression).

Overall, the discordance and the direction of the discordance were similar at both baseline and 12 months. Of the roughly 35% of patients with a change in NYHA class at 12 months, 20.9% showed improvement, and 14% worsening. Of the 75.1% who had a change of 5 or more points in KCCQ-OS, 48.3% had improvement and 26.8% had worsening.

There may be therapies that are not being initiated because there's not the full perception of the degree of impairment of that patient's health status. Gregg Fonarow

At both baseline and 12 months, older age and Hispanic ethnicity were associated with a higher likelihood of NYHA class being higher than the patient-reported KCCQ-OS score.

In landmark 12-month analysis, an improvement of 5 points or more in KCCQ-OS was shown to be an independent predictor of decreased all-cause mortality (P < 0.001) and the composite of all-cause death or HF hospitalization (P = 0.002). Improvement in NYHA class, however, was not associated with reductions in subsequent all-cause death, HF hospitalization, or the composite of death or HF hospitalization.

Physician Perception vs Patient Reality

According to Greene and colleagues, the finding that even patients who had changes of 10 points or more in KCCQ-OS score, whether it was improvement or worsening, generally had no corresponding change in their NYHA class, suggests that “limited change in clinician assessment of NYHA class would be consistent with other examples of clinician inertia in ambulatory HF care.”

To TCTMD, Fonarow said the study also shows that clinicians are missing opportunities to make a meaningful difference in their patients’ lives if they rely only on what they observe or what the patient tells them in office visits.

“Sometimes patients want to, in their brief interactions with clinicians, communicate things a little more positively, [because] their perception is that the physician wants to think that the treatment that they're offering is having a larger impact than maybe it is,” he added. “There may be therapies that are not being initiated, because there's not the full perception of the degree of impairment of that patient's health status.”

Likewise, Heidenreich points out that providing details on a form can take away the obligation some patients may feel to agree with their physician and not contradict what they are told about their symptom status and how they are faring overall.

Additionally, Heidenreich notes that the study builds on others, including one in which integrating patient-reported outcomes in metastatic cancer management was associated with better survival. Despite this, health systems have been slow to adopt the practice of using them. In recent years, he writes, only about one in five health systems regularly implement patient-reported outcomes, although as many as 70% say they plan to do so.